A time series analysis of disruptions to maternal and child health care in northern Togo during the COVID-19 pandemic in the context of an integrated primary care program

Public health crises, such as Ebola in West Africa from 2013 to 2016, can adversely affect the ability of communities to obtain routine health services, raising concerns about the impact of such crises on a wide range of health outcomes and trajectories [1, 2, 3]. There are concerns about the indirect impact of the COVID-19 pandemic on the health of women and children in sub-Saharan Africa due to disruptions in access and utilization of essential maternal and child health services [4].

Early predictions of the indirect effect of the COVID-19 pandemic estimated that disruptions in routine health care would be catastrophic and reverse the significant gains made over the past 2 decades in reducing maternal and under-5 mortality in low- and middle-income countries (LMICs) [5]. A survey carried out with Ministry of Health (MoH) officials from 105 countries in 5 World Health Organization (WHO) regions found that as of May 2020, there had been significant disruptions in a wide range of essential health services in nearly all countries, and that these reductions were reported to a greater degree in low-income countries [6]. Arsenault and colleagues conducted a study of 10 countries and found disruptions in health care that varied widely in both magnitude and duration [7]. A study examining disruptions in health-care services in 8 African countries between March and July 2020 found significant variation in the interruption of health-care service delivery across countries [8], indicating that there is a need to examine not only the extent of changes in health-care utilization but also contextual factors that mitigate or exacerbate such interruptions.

The capacity to maintain core functions and services during crises is a crucial component of health system resilience [9]. Declines in health-care utilization during epidemics and crises are driven by factors including reduced health care workforce, resource reallocation, strained supply chains, lockdowns, travel restrictions, financial hardship, and fear of exposure to infectious disease [10]. During the COVID-19 pandemic, several factors led to reduced health care use, including fear of contagion, loss of income leading to lack of ability to pay for services or transportation to care, prioritizing care for COVID-19 patients over routine care, redeployment of health-care workers to COVID-19 care, and lockdowns limiting travel and access to care [7]. Government and health system responses to crises vary across contexts and the strategies used can either lessen or amplify health-care utilization challenges [11].

On March 6, 2020, the first COVID-19 case was reported in Togo. After this, Togo experienced 3 distinct epidemic waves, namely in April 2021, August 2021, and January 2022. As of November 22, 2023, the cumulative count of confirmed COVID-19 cases in Togo stood at 39,524 and 290 reported deaths [12]. The objectives of this study were to examine disruptions in maternal and child health care during the COVID-19 pandemic in northern Togo, West Africa, and to explore community perceptions of COVID-19 and experiences seeking care during the pandemic. To our knowledge, this is the first study of health care use and perceptions of COVID-19 during the pandemic in Togo.

This study was conducted in the Bassar district in the Kara region, located in northern Togo, representing a total estimated catchment area population of 120,000 [13]. Togo faces significant public health challenges, including a high burden of preventable diseases, high rates of maternal and child mortality, and low utilization of the public sector health system [10, 14, 15]. Togo’s MoH has developed a national health plan to align policy with global best practices to reduce child mortality, yet Togolese children continue to die of low-cost treatable diseases [16]. In 2015, Togo’s MoH and a partnering nongovernmental organization launched the Integrated Primary Care Program (IPCP) in the Kozah district of the Kara region, followed by implementation in the Bassar district in 2018, Dankpen district in 2019, Kéran district in 2020, and Binah district in 2021 [17, 18]. The IPCP delivers evidence-based interventions, including (1) elimination of health-care costs for women and children under 5 years of age [19, 20, 21]; (2) proactive community-based care and delivery of reproductive and maternal health services by salaried community health workers (CHWs) [21, 22, 23]; (3) clinical mentoring at health centers [24, 25]; and (4) health center and supply chain operational improvements [26, 27, 28]. Together this package comprises the IPCP, a strategy that expands access to high-quality care, supporting the Togolese government’s universal health coverage targets [29].

After the first case of COVID-19 in Togo was reported on March 6, 2020 [30], the government issued social distancing guidelines, travel restrictions, and school and business closures. As of January 16, 2023, there have been 39,351 reported cases and 290 deaths from COVID-19 in Togo [30].

The programmatic response to the pandemic implemented by the IPCP, in partnership with the MoH, was focused on 3 objectives: (1) maintaining the health and safety of health-care workers; (2) continuing to provide essential, high-quality primary care to supported communities; and (3) supporting MoH partners in the pandemic response at a regional and national level. Particular attention was placed on increasing the use of personal protective equipment (PPE) by CHWs and health center providers, reinforcing infection prevention and control measures, and supporting physical distancing strategies during care delivery. These measures were of particular importance in assuring safety for health-care workers, a significant challenge in other LMICs where access to PPE was limited [31]. CHWs were provided training and equipped with educational materials on COVID-19 prevention for use during home visits and offered additional psychosocial support through stress management training.

This study focused on 5 public sector community-level health centers supported by the IPCP in the Bassar district, including Centre Médico-Social [CMS] Bangeli, Unité de Soins Périphérique [USP] Manga, USP Kabou-Sara, USP Koundoum, and USP Sanda-Afohou. In these communities, the IPCP supports 33 CHWs (6–11 per health center) who provide home-based proactive consultations to community members, serving a catchment area of approximately 40,000 individuals. This study also included health management information system (HMIS) data from 13 comparable (i.e., CMS or USP) health centers not supported by the IPCP for which adequate data were available: CMS Bikpassiba, CMS Kabou, CMS Sanda Kagbanda, USP Nangbani, USP Tchatchaminade, USP Baghan, USP Binaparba, USP Bitchabe, USP Bougabou, USP Byakpabe, USP Dimori, USP Kountoum, and USP Natchamba, serving a catchment area of approximately 80,000 people (Figure 1). The 18 total health centers represent all public health centers in the Bassar district.

In this study, we have adopted the Social Ecological Model (SEM) as a theoretical framework to examine disruptions in health care and to explore perceptions and experiences of community members during the COVID-19 pandemic in Togo [32]. The SEM is a theory-based model that considers not only the impact of multiple levels of influence on an individual’s behavior but also the interactions between those levels (Figure 2). The reasons for disruptions in health care use during the pandemic are multilevel, including individual fears of contagion, community health system failures, and policies like stay-at-home orders [33]. Our study examined disruptions to health care during the COVID-19 pandemic at 4 levels of the SEM: (1) organizational and health system factors, (2) community and social level factors, (3) interpersonal level factors, and (4) individual level factors. The study combines analysis of routine health center data, programmatic data collected through implementation of the IPCP, and in-depth interviews to explore the impact of COVID-19 on maternal and child health care in the Bassar district of the Kara region and to explore community perceptions.

This study used 3 data sources collected at multiple levels of SEM to examine health care disruptions and community perceptions and experiences of COVID-19 in the Kara region of northern Togo. These data include routine HMIS data, programmatic data from health centers, and qualitative data collected from community members (Figure 3). Explanatory mixed methods were used, with the qualitative results explaining findings from the quantitative analysis. Ethical approvals were obtained from the institutional review boards of the Togolese Ministry of Health in Lomé, Togo (ref: 017/2020/CBRS) and the Albert Einstein College of Medicine, New York, USA (ref: 2017–8411).

The District Health Information System (DHIS2) is a HMIS software used in 73 LMICs, including Togo, to collect and process health center data [34]. We extracted DHIS2 data for monthly health center-based maternal and child health services for the health centers referenced above for the period January 2018 to March 2021. Consistent with the WHO guidance on monitoring essential maternal and child health services during the COVID-19 pandemic [35], we examined indicators to assess changes in overall utilization of the health center as well as essential reproductive, maternal, and child health services. These indicators included total number of health center consultations; children’s first, second, and third doses of DTP-HepB-Hib (diphtheria, tetanus, pertussis; hepatitis B virus; and Haemophilus influenza type B) vaccination; women’s health center births; number of women seen for the first and fourth antenatal care visits; number of pregnant women receiving HIV counseling; and the number of injectable and implant contraceptives for women distributed through the health center. Injectable and implant contraceptives are the method of choice for most Togolese women who use modern family planning [36]. The number of pregnant women receiving HIV counseling was considered a proxy for HIV testing as the number of HIV tests administered to pregnant women was not available in the HMIS data. For each outcome, the number of visits or services delivered was summed across all IPCP-supported health centers and health centers not supported by the IPCP. The indicators chosen for the analysis represent key performance indicators for assessment of the IPCP and the government of Togo. These data are routinely collected and are accessible.

Programmatic data on CHW home visits and timeliness of child treatment is collected during implementation of the IPCP. CHWs complete paper forms in the field and submit them to both IPCP staff and health center managers for monthly review. Health center managers add consultations conducted by IPCP-supported CHWs to DHIS2 data along with other data collected from CHWs in the region. For the present study, programmatic data were collected from the 5 IPCP-supported sites to measure the number of CHW consultations and proportion of cases of child illness treated by CHWs within 72 h per month.

Qualitative data were collected through semi-structured interviews with stakeholders in December 2020, 9 months following the start of the pandemic in Togo. Participants were selected using purposeful sampling [37] and included community members (n = 10), CHWs (n = 5), and health center managers (n = 5) from the 5 health centers supported by the IPCP. One CHW, 1 health center manager, and 2 community members were selected from each health center (Figure 4). This sampling method allowed for achieving diversity of participants in terms of gender, age, and occupation (Supplemental file 1). All participants had been in the area for at least the prior year. Semi-structured interview guides were developed based on the SEM [32] to assess factors influencing changes in health-seeking behaviors (Supplemental file 2). Data were collected in parallel from different respondent groups and then integrated at analysis [38]. Two Togolese qualitative researchers (DD, EM), one with a PhD in Sociology and the other with a bachelor’s degree and both having extensive experience conducting semi-structured interviews, conducted all interviews in-person and in the preferred language of the interviewee (French or Bassar). Interviews were audio-recorded, transcribed verbatim, spot checked for accuracy, and translated to English, according to standard protocols [39].

We used an interrupted time series model to estimate changes in health-care service utilization during the COVID-19 pandemic. Interrupted time series analysis utilizes a time series of a particular outcome of interest to establish an underlying trend, which is “interrupted” by an intervention at a known point in time; changes in level and gradient of the trend in outcomes are then compared to the expected trend in the absence of the intervention (i.e., the counterfactual scenario) [40]. We estimated both level and trend changes for each type of visit, with analyses stratified by IPCP-supported sites and sites where the IPCP is not being implemented, using the following model:

in which Y_t is the number of visits in month t, β₀ is the intercept, β₁ is the level change in the outcome associated with the start of the pandemic, β₂ is the trend in the outcome in the pre-pandemic time points, β₃ is the difference in trend in the outcome in the post-compared to the pre-pandemic time points, γ_k represents a fixed effect for month k, and ε_t is the error term for month t. The model assumes that the pandemic began in March 2020, represented by T₀.

For missing values of health-care visits, we used linear interpolation between neighboring time periods to impute the missing values. Any sites that were missing the majority of outcome values were dropped from the analysis for that particular outcome. To assess model fit, we evaluated the distribution of the residuals to ensure they were approximately normal and used plots of the autocorrelation function to verify there was no residual autocorrelation.

Routine programmatic data from 2019 to 2020 on the total number of CHW visits and the timeliness of treatment of child illness by CHWs (i.e., treated within 72 h) were reviewed to provide additional context regarding the continued provision of community-based care during the COVID-19 pandemic. We used 2 sample T-tests to estimate changes in programmatic measures before and after the COVID-19 pandemic, while maintaining the assumption that the pandemic began in March 2020. Any missing values were excluded in the comparison of sample means for each measure. Estimates for pooled T-tests were reported after confirming the equality of variances between samples.

The data were analyzed using a rapid qualitative approach [41, 42, 43]. Three qualitative researchers (JH, EZF, DD) developed and piloted a summary template based on the SEM domains (Supplemental file 3). Each interview transcript was then independently summarized by 2 of 3 researchers to avoid lone researcher bias in qualitative analysis [44]. Each transcript summary was compared and discussed by 2 researchers and discrepancies were resolved by consensus [45]. Data from the summaries were added to matrices arranged by interviewee type (i.e., community member, CHW, health center manager) and SEM domain [45]. Other segmentations of the data (e.g., age, income, education) were applied as needed to contextualize the findings. For each SEM domain, analytic memos were drafted summarizing major themes and findings.

We observed a temporary decrease in the number of people seen in IPCP health clinics for any reason at the start of the pandemic in March 2020 (−547.7 expected monthly visits, 95% confidence interval [CI] −1085.5, −9.9). The drop in visits at the sites was temporary and by summer 2020 the number of visits among the health centers increased, with a slightly higher trend throughout 2020 compared to pre-pandemic (trend difference = 6.6, 95% CI −7.6, 20.8) (Figure 5). In contrast, there was a slightly declining preintervention trend in total visits at health centers not supported by the IPCP, and no significant difference in expected monthly visits following the start of the pandemic in March 2020 (level difference = 77.2, 95% CI −440.3, 594.7), with a slightly higher trend throughout 2020 compared to the pre-pandemic (trend difference = 8.8, 95% CI −4.9, 22.4) (Figures 5–7).

At IPCP-supported health centers, small reductions in the number of visits at the start of the pandemic were observed for first and fourth antenatal care visits (−22.3 [95% CI −57.7, 13.0] and −19.7 [95% CI −36.6, −2.8] expected monthly visits, respectively), though only the change in fourth visit reached statistical significance. There were no significant trends in monthly visits for first antenatal visit (0.7 [95% CI −0.3, 1.6]) and fourth antenatal visit (−0.1 [95% CI −0.5, 0.4]) following the start of the pandemic at IPCP health centers (Figures 6 and 7). Similar findings for first antenatal care visit were observed at the health centers not supported by the IPCP (Figures 6 and 7). In contrast, the number of fourth prenatal care visits increased significantly at sites not supported by the IPCP at the start of the pandemic (level difference = 58.4 expected monthly visits, 95% CI 44.8, 72.1), which was then following by a substantial decrease, such that the trend difference was −1.0, 95% CI −1.4, −0.6 in the post-pandemic compared to the pre-pandemic period. This increase was due in part to a notable increase in fourth prenatal care visits at USP Bitchabe beginning in March 2020 and continuing through January 2021.

At sites not supported by the IPCP, there was a slight temporary drop in the number of pregnant women receiving HIV counseling (−20.7 [95% CI −53.1, 11.6]), followed by an observed increase in the post-pandemic trend (trend difference = 1.2, 95% CI 0.4, 2.1). These changes in the expected number of pregnant women receiving HIV counseling were not observed in IPCP-supported health centers (Figures 6 and 7). There were no significant changes in either the levels or the trends in the number of services delivered in the pre- versus post-pandemic periods in number of vaccinations, injectable and implant contraceptives distributed, or births at both IPCP-supported health centers and centers not supported by the IPCP.

Analysis of IPCP programmatic data indicated there were no meaningful changes observed in the monthly counts of CHWs visits following the pandemic start in March 2020 (P = 0.3). Prior to the COVID-19 pandemic, we observed an average of 3,960 CHW visits per month in all IPCP-supported health centers in Bassar, with a range from 3,097 to 5,556 visits per month during this time. Meanwhile, there were 4,282 CHW visits per month, on average, after the COVID-19 pandemic, with visits ranging from 3,106 to 5,887 per month. We observed expected increases in CHW visits during rainy season months (July and August) in both time periods. Additional analysis determined that the timeliness of treatment of child illness by CHWs increased after the start of the pandemic (P < 0.0001). Before March 2020, the health centers treated between 89% and 99% of children within 72 h per month compared to 97%–100% of children after the start of COVID-19. This represents an average of 95% of children treated on time per month before compared to 99% after COVID-19.

Several multilevel factors influencing health-care utilization during the first year of the COVID-19 pandemic were identified in the qualitative data collected from community members.

Consistent with the quantitative findings, interview participants reported observing continued utilization of health-care services during the pandemic. One exception was when health center managers described temporary drops in consultations due to travel restrictions in place at the start of the pandemic.

Participants described increases in care seeking at the health centers due to concerns about COVID-19, which reflects effective community education about the risks of exposure and confidence in safety protocols.

Participants reported regular and clear communication about the COVID-19 pandemic, including government and private sources (e.g., radio) and community-based sources (e.g., health centers, village chiefs). CHWs played a central role in convincing the community that COVID-19 was real and communicating about risks and prevention strategies. Participants were aware of the risks of COVID-19 and methods to prevent transmission (e.g., hand washing, social distancing, masks).

In response to the pandemic, health centers instituted social distancing and PPE protections, including mask requirements and social distancing. Health center managers described the importance of the newly installed handwashing stations for infection control.

Health center managers described challenges encouraging compliance with COVID-19 precautions while providing care.

CHWs received PPE and were trained to follow social distancing guidelines. CHWs continued to provide care in their communities, with consultations occurring outside the patients’ homes, which was especially important to patients afraid to go to the health center during the pandemic.

CHWs served an important role providing care for those who felt uncomfortable seeking care at the health center. However, some CHWs expressed concern about how social distancing impacts their interactions with patients, in particular, not being able to get as close to patients at their homes.

During the pandemic, CHWs conducted consultations outside whenever possible and at a distance, which may have increased the chances of being overheard by others. CHWs were concerned about how social distancing might affect patients’ confidence in sharing personal health concerns, for fear of being overheard by neighbors or other household members.

Health center managers and CHWs partnered with community leaders and authorities to limit the spread of COVID-19 through community gatherings and traditions. Health center managers and CHWs discussed engaging community leaders and village chiefs to help limit large community gatherings and limit the risk of exposure.

This study has examined disruptions in health care during the COVID-19 pandemic and explored community perceptions and experiences of health care use in Togo, West Africa. Using the SEM, this study found that while the COVID-19 pandemic resulted in some disruptions in health care use, continuity of care increased to normal levels within a short amount of time. It should be noted that mortality due to COVID-19 was low in Togo, which may have influenced perceived risk and sustained health-care utilization. Others have found that during the COVID-19 pandemic, disruptions to health care were not driven by COVID-19 severity, and that the impact of the pandemic on health care use was felt everywhere [7]. Even places with low incidence, like Togo, saw disruptions to health care due to travel restrictions, fear of contagion, and social distancing requirements that limited access to care. However, our study found that these disruptions were short-lived and did not significantly impact health care use.

While there were some differences in the expected number of monthly visits at the start of the pandemic and differences in trends throughout the latter part of 2020 compared to the pre-pandemic periods, we did not observe any consistent patterns in health center service utilization changes coinciding with the COVID-19 pandemic. Though there were some differences in pandemic-related changes in service delivery across IPCP-supported health centers and those not supported by the IPCP, there were no substantial sustained pandemic-related shifts in health-care utilization in both types of centers. These observations in the health center data were supported by the qualitative data from the IPCP-supported health centers indicating that patients continued seeking care from health centers.

Several factors identified in the qualitative data help to explain the sustained use of health-care services during the COVID-19 pandemic. Interview participants described extensive risk communication efforts by trusted community sources, the use of PPE and adoption of risk minimization strategies by health center staff and CHWs, and adaptations to health care delivery that enabled the continued provision of care at health centers and in the community. Other studies have highlighted the importance of access to information through telecommunications and trusted sources during the COVID-19 pandemic [46]. The health system in Togo has demonstrated the resiliency associated with an adequately trained health workforce, the implementation of strong infection prevention and control measures, and communication by trusted community sources [9]. These strategies were adopted swiftly following the first reported cases in Togo in March 2020, despite the low number of total cases throughout the country.

There is evidence that the impact of the COVID-19 pandemic on the delivery of routine health care has been variable across health services and geographic regions [7]. A study examining changes in health-care utilization across 8 countries in sub-Saharan Africa (Cameroon, Democratic Republic of Congo, Liberia, Malawi, Mali, Nigeria, Sierra Leone, and Somalia) found that all countries experienced drops in outpatient consultations and most experienced drops in vaccination of children following the start of the COVID-19 pandemic, but there was substantial variability in the observation of any reductions and extent of disruptions in family planning, antenatal care, institutional deliveries, and postnatal care across these countries [8]. Fejfar and colleagues observed significant sudden drops in outpatient services, but less consistent changes in other services including family planning, across 7 countries [47]. Studies in individual countries have found declines in specific services; for example, a study conducted in Rwanda observed significant declines in antenatal care, health center deliveries, postnatal care, and vaccinations during the early stages of the COVID-19 pandemic [48]. An assessment of global and regional disruptions to vaccination coverage of the third dose of a diphtheria-tetanus-pertussis vaccine (DTP3) and the first dose of a measles-containing vaccine (MCV1) during the COVID-19 pandemic in 2020 found that global disruptions were most severe in April 2020, with delivery improving and approaching expected levels in May to December 2020 [33]. A study of changes in childhood vaccination rates in 4 countries during the pandemic found widely varying patterns, with one country having much lower rates [49]. Another in Burkino Faso did not observe the expected overall low childhood vaccination rates predicted in LMICs, however, results varied by geography, exacerbating regional disparities [50]. Among the Global Burden of Disease world regions, sub-Saharan Africa was observed to have the smallest absolute declines in vaccination coverage attributable to the COVID-19 pandemic.

COVID-19 presents a challenge for continuity of primary care, particularly in underserved and rural contexts [51]. A study examining disruptions in maternal care in 6 countries found that despite variations in context and differing degrees of disruptions, all health systems shared the need to adapt to the challenges brought by the COVID-19 pandemic, including the use of demonstrated strategies that can be adapted to the local context [11]. CHWs can play an important role, having established trust in the communities they serve, and enabling access to health care through home visits and education about disease outbreaks [31]. While existing CHW programs have been demonstrated to assist in pandemic response [52, 53], adding additional tasks to CHWs’ workload (e.g., contact tracing, building awareness) can reduce the number of consultations CHWs can complete and in some circumstances can lead to social ostracization and increased vulnerability [52, 54, 55]. Programmatic data from the IPCP indicates that CHWs were able to continue the provision of care throughout 2020, without any drop in total monthly number of home visits. In addition, data from similar organizations in Kenya, Uganda, Mali, and Malawi found that there has been continued community-based provision of primary care services by CHWs throughout the pandemic [56]. When engaging CHWs in pandemic response, adequate training, support, and protections are critical for ensuring their safety.

While some early predictions of the COVID-19 pandemic expected a devastating impact on the health of low-resource and rural communities [5, 57], others highlighted the successful initial response of countries in sub-Saharan Africa [58, 59]. It has been acknowledged that multiple epidemics and natural disasters have built collective knowledge and prepared governments, health systems, and communities to respond quickly and appropriately to global threats [60]. Importantly, a focus on African assets (i.e., community-based health care, youth engagement, and social capital) could help accelerate responses to future pandemics and crises [61]. Various protective factors may be contributing to mitigating the impact of COVID-19 in Togo, including a history of infectious disease epidemics [59, 62], quick government action to impose travel restrictions and contact tracing [63], cultural factors including social capital and collectivism [61, 64], and preestablished community-based health-care systems [65]. However, the trajectory of the COVID-19 pandemic has been distinct in sub-Saharan Africa, and the ability of these protective factors to continue to mitigate the impact may be minimal given the slow update of COVID-19 vaccines in Africa [66] and the emergence of new and more contagious variants that lead to greater infection and severity of disease [67].

This study utilized routine health center data to examine changes in the utilization of essential maternal and child services during a global pandemic. The monthly collection and reporting of these programmatic data allowed for the use of a robust quasi-experimental method, interrupted time series analysis, to quantify the differences in service utilization following pandemic start in March 2020, as well as changes in trends of utilization through 2020. Given that we used a linear model, we checked the residuals for normality and found them to be sufficiently normal [68]. Residual autocorrelation would occur if there were time trends in the residuals, suggesting the model had omitted variables related to time that would need to be included. When we checked the residuals for autocorrelation, we did not find any. In addition, the collection of qualitative data was central to better understanding local challenges and adaptations to care delivery during the COVID-19 pandemic. Data collected at multiple levels of the SEM serve to validate findings, ensure richness in the data, and elucidate differences in perspectives given differing roles [45]. In this explanatory mixed methods analysis, the qualitative data served to help explain the quantitative results.

Our study had several limitations. This analysis was limited to a single district in the Kara region of northern Togo; additional analyses are required to determine if these results are generalizable to northern Togo or the country as a whole. In the context where most administrative data continue to be collected by paper and then input in the DHIS2 system, health center data may be limited by reporting errors or bias, outdated or incorrect population estimates, and changes in indicator definitions [35]. Additionally, data are reported by month, which could mask decreases in utilization in a single month while still showing an overall rate that is increasing. There are not sufficiently accurate estimates of non-IPCP supported site catchment areas available, therefore, we could not account for population size in the analysis. While interviews were conducted with health center managers and CHWs from all 5 IPCP-supported centers included in the analysis, the study did not include individuals from sites not supported by the IPCP. Interviews were conducted during a short time window compared to the longer time period of the study, which limits findings to participants’ reflections and experiences up until that point. In addition, interview participants are segmented by gender, as almost all health center managers in Togo are men and all of the CHWs in our sample are women. This segmentation could have influenced the responses due to gender differences in the samples. The relatively small number of time points following March 2020 and the smaller number of counts for the services provided in one district (compared to national estimates, for example) may have limited power to detect changes.

This study found that in one region in Togo, despite travel restrictions, fear of contagion, and other factors influencing health care use, only short-term disruptions in health care were observed during the first 9 months of the COVID-19 pandemic. While disruptions to health care have been observed globally, the extent of these disruptions vary widely, and there are no clear patterns between low- and high-income countries to explain these differences [7]. In the context of this study, an integrated health system, including well-equipped health centers and CHWs who focus on both COVID-19 prevention and routine provision of care, could have supported continuity of care during the pandemic [65]. Community-based primary care is an essential part of building trust in the health system in many parts of Africa and continues to be an effective strategy for both routine health care and resiliency in response to larger health crises [61, 63]. Countries, regardless of income level, can prepare for future public health crises by supporting the development of strong primary care including integrated community-based health systems and engaging health center staff, trained CHWs, and trusted community leaders in emergency preparedness and response.

The quantitative data underlying this article were provided by permission of the Togolese Ministry of Health and will be shared on request to the corresponding author with approval by the MoH. The qualitative data underlying this article will be shared on reasonable request to the corresponding author.

The supplemental files for this article can be found as follows:

Supplemental file 1. Demographics of semi-structured interview participants (n = 20).

Supplemental file 2. Semi-structured interview guides.

Supplemental file 3. Social Ecological Model transcript summary template.

The authors acknowledge and thank the Togolese Ministry of Health, including health authorities in the Kara region and the district of Bassar, and Integrate Health/Santé Intégrée staff in New York and Togo for their efforts in developing and implementing this study. They would like to extend their gratitude to the health center managers, CHWs, and community members who participated in interviews, whose contributions are invaluable to this work.

Research reported in this publication was generously supported by the Robert J. Havey, MD Institute for Global Health’s catalyzer fund at Northwestern University, Feinberg School of Medicine (grant #1041).

None declared

Jessica Haughton led the conceptualization of the work, funding acquisition, data analysis and interpretation, and writing the original draft. Désiré Dabla led data collection, participated in data analysis and interpretation, and reviewing and editing the manuscript. Dana E. Goin led analysis and interpretation of the quantitative data. Amanda W. Singer participated in quantitative data analysis, interpretation, and writing sections of the paper. Elissa Z. Faro led qualitative data analysis and reviewed and edited the manuscript. Essodinam Miziou led quantitative data collection and supported qualitative data collection. Assiongbonvi Kangni-Zovoin, Sibabe Agoro, and Didier K. Ekouevi reviewed and edited the manuscript, providing insight on context of the study. Kevin P. Fiori and Lisa R. Hirschhorn participated in the conceptualization of the work, funding acquisition, and editing of the final draft.

How to cite this article: Haughton J, Dabla D, Goin DE, Singer AW, Faro EZ, Levano S, Miziou E, Kangni-Zovoin A, Agoro S, Ekouevi DK, Fiori KP, Hirschhorn LR. A time series analysis of disruptions to maternal and child health care in northern Togo during the COVID-19 pandemic in the context of an integrated primary care program. Adv Glob Health. 2024;3(1). https://doi.org/10.1525/agh.2024.2123937

Editor-in-Chief: Craig R. Cohen, University of California, San Francisco, CA, USA

Senior Editor: Andres G. Lescano, Cayetano University, Peru

Section: Improving Health and Well-Being